1994_Entomol_Probl_Gabel_et_al_25_1

8/8/2019 1994_Entomol_Probl_Gabel_et_al_25_1

1/7

Entomol. Probl., 25(1): 1-7, 1994. ISSN 0071-0792

O l f a c t o r y r e s p o n s e s o f Lobes ia bo t r ana fe:m.ales ( Lep i d o p t era : To r t r i c i d a e ) to Tanace turn vu lga re ( A s t era cea )

f lo-wer e x t r a c t s a n d f r a c t i o n s Bruno GABEL1, Frederic MARION-POLL2, Vaclav SUCHy3,Roger ROEHRICH\ Peter HRADSKYs & Denis THIERY

'Research Institute of Viticultu re and Enology, Matuskova 25, 83311 Bratislava, Slovakia'INRA, Station de Phytopharmacie, Route de Saint-Cyr, 78026 Versailles Cedex, France3Pharmaceutical Faculty, Department of Pharmacognosy and Botany, 83232 Bratislava, Slovakia

'INRA, Station de Zoologie, BP 81, 33883 Vilienave-d'Ornon Cedex, FranceSState Forest Products Research Institute, Lamacska cesta 1, 83330 Bratislava, Slovakia Laboratoire de Neurobiologie Comparee des Invertebres, INRA-CNRS (URA 1190), BP 23,91440 Bures sur Yvette, France

Abstract. The European grapevine moth, Lobesia botrana DEN. ET SCHIFF. is a phytophagous speciesrestricted to a narrow range of host-plants. During the summer when tansy occurs in its ecosystem,females can strongly be attracted to tansy flowers. The ecological value of such an attraction is unclear,since tansy reduces the moth's mating and oviposition and is toxic for larvae. We used electroantennogram recordings and olfactometer bioassays to clarify the responses of L. botrana females totansy flower odours. Our data demonstrate that tansy odour is well detected by females, and the resultantEAG responses are dose dependent. The largest magnitude EAG responses were elicited at each doseby the two extracts of tansy (EO and E30). When presented a dual choice of a test substance vs solventcontrol, more than 80 % of the females were attracted by either of these two extracts of tansy flowers,while singly the three fractions attracted between 66 % and 56 % of females. The rate of attractionincreased on the second and third days of the experiment, when females are 4 to 5 days old. Thiscorresponds to prior field observations. A positive correlation was found between the EAG and behavioural responses, suggesting that EAG analysis can usefully screen out biologically relevant odours. Ourresults yielded information concerning odour extraction and fractionation procedures. It might impact thepractical use of tansy odours in grapevines protection.Key words. Lobesia botrana, tansy, flower extract, attraction, plant odour, electroantennogram,olfactometer.

Introduction

Plant diversity is an important feature of the environment of phytophagous insects, even if the agroecosystem structure tends to reduce this diversity. There arenumerous examples of weeds that play important roles in the biology of pest insects(PERRIN, 1977; TAYLOR, 1977). Mixed stands of plants can affect the development ofphytophagous insect populations by hindering successive patterns of host-plant selection, like searching behaviour (UVAH & COAKER, 1984; THIERY & VISSER, 1986), duration ofthe periods of activity, and oviposition behaviour (STANTON, 1982; LATHEEF & ORTIZ, 1984).European grapevine moth (EGVM), Lobesia botrana DEN. ET SCHIFF., is a seriouspest of european vineyards, despite its limitation to a narrow range of alternative hostplants (STOEVA, 1982). Tansy, Tanacetum vulgare L. is an Asteraceae weed species thatnaturally occurs in many agroecosystems, including vineyards when no herbicide isapplied. Tansy flowers strongly attract females of EGVM (GABEL, 1992). This attraction Entomological Problems

1


2/7

can be mimicked under controlled conditions with synthetic blends of monoterpenes,that have been identified in the odour of flowering tansy (GABEL et aI., 1992). Theecological value of such an attraction is, however, still unclear, since it has beendemonstrated that tansy is not a host plant of EGVM. Further the flower odour of tansyhas recently been shown to strongly reduce oviposition and also mating activity (GABEL& THIERY, 1994) which may prevent offspring from having to face the toxic allochemicalsof tansy (LUSTNER, 1914).

The aim of the present research was to define simple and low cost extraction/fraction procedures leading to efficient odourant baits which could be proposed forpractical using in grapevine protection, and to determine the behavioural activity offemales to odourants as a function of female age. The biological activity of the variousextracts and fractions has been assesed by two different methods: electroantennogram(EAG) recordings and behavioural olfactometer bioassays.

Materials and methods

Plant extracts. Tansy flowers (without their pedicels) were collected in a Vineyard located inModra (south west of Slovakia) between the end of july to mid-August. Extractions were made on freshmaterial, immediately following the harvest.Essential oil of tansy: 500 g of flowers were water steam distilled during two hours using adistillation-extraction device modified after Likens and Nickerson (PHARMACOPEA BOHEMOSLOVACA, 1987).Half milliliter of essential oil (EO) was obtained (equiv. to ca 10 g of flowers/ml).Fractions of tansy flowers: 1000 g of flowers were macerated during 48 hours in 2000 ml of pureethanol and than filtered through the cellulose. Two batches of 650 ml were evaporated until the ethanolsolvent was removed using a vacuum rotary evaporator. The dry extracts were then diluted each in 1000ml of hot purified water and distilled by water steam using the device described above until obtention of:a) a volume of 30 ml (sample E30; equiv. to ca 10 g of f1owers/ml); b) 3 consecutive fractions of 50 ml eachcorresponding to different volatilities (samples F50-1, F50-2 and F50-3; equiv. to ca 2 g of flowers/mleach).

Insect material. The insects originated from a stock culture (INRA, Bordeaux, France). that wasreared on semi-synthetic diet and annually infused with wild insects. Females used for bothelectroantennograms and bioassays were 2 days old and mated. Behavioural experiments were conducted under 14L18D, light period being followed by artificial sunset (constant decrease from 1200-1250to 20-30 lux during one hour) and night period followed by sunrise (inverse). Temperature (22C 1C)and relative humidity (60 % 5 %) were held constant.

Electroantennogram set-up. Electroantennogram (EAG) measurements were conducted onhead-antenna preparations. Electrodes were glass capillaries with an Ag/AgCI wire inserted and filledwith physiological saline solution (KAISSLlNG, 1987). The recording device consisted of an input probe, aCarrack PB 181-11 preamplifier, a Tektronix 5223 differential oscilloscope and a microcomputer usingcustom software adapted from MARION-POLL & TOBIN (1991). Preparations presenting impedences higherthan 15 MOhm were discarded. The antenna was continuously bathed with purified and humidifier air(4.67 ml/sec). Odourants were placed, without dilution, onto 350 mm' of filter paper inserted into a 150mm long Pasteur pipette (see amounts beneath). During stimulation (0.82 s) the odourant air from pipette(2.19 ml) was injected into the clean air flow and then blown over the prepared EGVM antenna. Four litand 40 li l of a solution of hexanol diluted in paraffin oil (10-' vollvol dilution) were used as standards.Responses to EO were recorded from 20 females, using the following doses: 1lil, 4 lil, 8 lil, 16 li l and40 lil. Series of stimulations including the three different fractions have been performed on 10 femaleswith two doses (4 li l and 40 lll). The stimulation sequence was: 2 successive standards, 2 fractions,standard, 2 fractions, standard, etc .. Each EAG response, was expressed as a percentage of the valueobtained with the hexanol standard (corrected according to the decrease observed with sequentialstimulation by hexanol). EAG were then expressed as a mean SD of the recorded percent responses.

Bioassay. Behavioural activity of odour substances was measured in an olfactometer designedspecially for EGVM females. It consists in a 10 L glass barrel, with two traps (100 ml glass jar with 5 mmI. D. hole made in the lid) hung at a height of 25 mm above the bottom. Purified and humidified air wasblown in from the bottom of the barrel and exitted from the top, providing a constant and vertical air flow(625 ml/min, speed: 0.5 m/sec). Traps contained 26 ml of purified water with 2 drops of wetting agent(Citowett; BASF Germany), one short glass testtube (10 mm I. D.) filled with 100 ml of paraffin oil (usedas releaser) and only in odourant trap 400 ml of tested blends. Ten females were released into the barrel

2


3/7

per test, between 10 and 14 test repetitions were made for each fraction.The number of females caughtwas recorded for each trap every morning over a 3 days span.We calculated a daily differential trapping rate (TR, in percent) between control and odouranttraps: TR = 100 x (0 - C) / (n - (m+t)), where a is the number of females caught in odourized traps, C the

number of females caught in control traps, n the total number of females, m the number of females founddead and t the number of females trapped on previous the day. The mortality observed during theexperiment was equivalent for the 5 tested blends, varing within 2- 7 % of the individuals released.

Statistics. Statistical differences between the evoked EAG as well as the final behavioralresponses to tansy extracts and fractions were evaluated by t-test statistics (WEIR, 1960). Fourfold tablesprocedure was used to compare the trapping rates, while relationships between EAG and olfactometerresponses were examined by a linear correlation analysis (SACHS, 1984).

ResultsEAG responses to th e various fractions. The mean value of the responses to

the hexanol standard were 1.65 0.30 mV (4 Ill) and 2.06 0.54 mV (40 Ill). The meanEAG response to EO increased as the dose deposited on the filter paper increased, butno significant differences in EAG magnitude were found between the dose levels. Themean response to EO (relative to hexanol standard at 4 III dose) increased from234 59.6 % at the dose 1 III to 284.5 90.9 % at the dose 16 III (Figure 1). The dose of40 III provoked "hyperpolarizations" (i. e., a positive polarity EAG potential) in allrecorded antenna. The highest mean EAG response to fractions were recorded withstimulants of 40 III of E30 (4.75 0.76 mV) corresponding to 273.1 % to standard at thedose of 40 III and the lowest percent EAG response (79.6 %) with 4 III of the fractionF50-2 (1.28 0.37 mV) (Table 1). At both of the two doses used the highest EAGresponses were obtained with EO and E30 (t-test, P < 0,05). Fraction F50-1 released alower response as compared to E30 at the same doses (Hest, P < 0,05). The two otherfractions (F50-2 and F50-3) released the lowest responses at both doses (different fromF50-1 at P < 0,05, t-test). Sample F50-1 at 40 III dose elicited EAG responses of astatistically equal magnitude to those elicited by the EO and E30 samples at 4 III dose(Table 1).

350% -y--------------------------------,

300%

250% I I 1 I200% ->E150%

IIJ.I 4IJ.I 8IJ.I 16 IJ.I 40 IJ.I

Fig. 1. Electroantennogram (EAG) responses of EGVM mated females to different doses of tansyessential oil (EO). Responses are expressed as a percentage of the response to standard (4 J.l1 ofhexanol in paraffin oil 10-2 vlv). Open-squares represent the mean value from 20 individuals, thin barsrepresent standard deviation. The 40 J.l1 dose stimulation is not plolted because it resulted in a reversedpolarity of the potential (i. e. hyperpolarization).

I

3


4/7

I _ EO o E30 FSO-1 FSO-2 II!lI FSO-3!80% (a)

60%

40%

20%

0%3 days 4 days 5 days

Fig. 2. Daily attractancy indexes (%) of EGVM mated females to extracts (EO, E30) and fractions(F50-1, F50-2, F50-3) of tansy flowers as a function of moths age. Index is calculated from femalesremaining each day in the barrels. Within an age class, different letters indicate statistical differences atP < 0,05 (fourfold table procedure).

Correlation between EAGs and attractancy30 0

E 3 0 ~ * 250 r : O.!J98

EO0 20 0 F50-1 x E30 r : 0.996

F50-2150 F50-3100 /50-2 ...FSO-31!j1 -elIE 4 p.-l40 p.-l

5050 60 70 80 90 100

Attractancy [%]Fig. 3. Correlation between EAG relative responses and accumulative attractancy (after 72 hours) inLobesia botrana for two different EAG doses of tansy odours. EO and E30 are raw extracts, others aredifferent fractions. Data obtained with EO (40 ~ I was discarded because of hyperpolarization.

4


5/7

Table 1. EAGs (% of standard) and percent accumulative attractancy (after 72 hours) of mated femalesEGVM to tansy volatiles. EAG and behavioral responses are expressed as mean values(standard deviation). EAG standards were 4 ~ and 4 0 ~ 1 doses of hexanol 10- 2 , for respectivetest doses. Similar letters indicate no statistical differences at P < 0,05 (t-test).

ProductEOE30

F50-1F50-2F50-3

4 ~ EAG

40 ~ II

Attractancy I

217.9 (12.4) b198.2 (15.2) b

hyperpolar.273.1 (16.3) a

89.6 (0.9) a'85.7 (1.2) a'

109.2 (3.3) c79.6 (4.9) e81.6 (2.5) e188.6 (5.4) b153.8 (2.7) d149.7 (3.9) d

66.4 (1.5) b'55.8 (2.7) b'56.1 (2.3) b'

Behavioural responses. The extracts EO and E30 attracted significantly moremated females (fourfold tables procedure, P < 0,05) than each of the three fractions ofE30 sample (Table 1). The percentages of mated females caught per day were 24 %,49 %, 74 %'for EO and 24 %,47 %,65 % for E30.We found an effect of age on the behavioural response. The number of femalestrapped by the two most attractive blends (EO and E30) significantly increased on the2nd and 3rd day of experiment which correspond to 4 and 5 day old females (fourfoldtables procedure, P < 0,05) (Figure 2).Are EAG responses and attractiveness correlated? Most of EAG experimentssuppose the implicit assumption that large magnitude EAG responses are released byvolatiles of behavioural importance (either stimulation or inhibition) (VISSER, 1979; MAYERet aI., 1987). We have checked whether such an assumption is valid in L. botrana andwhich from two series of EAG responses is the best correlated with the attractancy.Significant linear correlations where found between the attractiveness and the relativeEAG amplitudes for each tested blends (Figure 3). We can conclude in this firstappraisal, that recording L. botrana EAG responses to volatiles may be used for a rawestimation of attractiveness.

Discussion

The present results show that olfactory receptors of . EGVM females detectvolatiles produced by tansy flowers. Significant EAG responses are observed with dosesas low as 4 ~ of extracts, which correspond to about 40 mg equivalent of tansy flowers(TFE). The amplitude of the EAG responses recorded with the essential oil of tansy andwith the various fractions are always greater than EAG responses obtained with extractsof various parts of its grapevine host plant (8. GABEL, unpublished data). The largeolfactory stimulation elicited by tansy volatiles can be explained by the high sensitivity(or high numbers of olfactory receptors) of EGVM to monoterpenes as demonstrated bycoupling GC-EAD recordings (GABEL et aI., 1992). High doses of essential oilreproducibly provoked positive polarity "hyperpolarizations". This is probably due tochemoreception of one of the three isomers of thujone, which combined represent themajor constituents of tansy essential oil (P. HRADSKY, unpublished data) like in DEMBITSKIet al. (1984). The global extract E30, obtained by a different extraction procedure, didnot cause the same effect. The 3 fractions from E30 released lower EAG responses ascompared to E30 and EO extracts. This is partly due to lower TFE in these fractions

5


6/7

(2 g/ml TFE with fractions vs 10 g/ml TFE with extracts). However, since fractionscorresponding to 80 mg of TFE (40 ~ of F50-1, F50-23, or F50-3) released lower EAGresponse than 40 mg of TFE for EO or E30 extracts (4 ~ I ) , we hypothezise that lowerresponses to each of the three fractions are the result of reduction in mixture complexity.This is supported by comparisons of GC patterns on major compounds: we found in eachof the three fractions (F50-1, F50-2, F50-3) approximately two times fold less compounds than in E30 (P. HRADSKY, unpublished data).

The two global extracts (EO and E30) also provoke a strong attraction of females.After 3 days of experiment 89.6 % and 85.7 % of females were recovered in odouranttraps baited with EO and E30, respectively. The three fractions of E30 are less attractiveto females. From these fractions, F50-1 which was obtained by the first collection of thedistillation (higher concentration of volatile substances measured by GC analysis), is themost attractive (as compared to F50-2 and F50-3). This fraction (F50-1) relesed alsohigher EAG responses as compared to F50-2 and F50-3.

Our present data also reveal significantly higher attraction with all odourants onthe second and third days of experiments. This corresponds to the optimal age offemales collected in the field, females about 4-5 days old (mated and having alreadyoviposited) being the most abundant on tansy flowers (GABEL, 1992). We cannotphysiologically explain the effect of the age on the increased responsiveness. We couldnot exlude sensitization (result of sustained exposure to odour) to partly interfere withthe behavioural observation. Associative learning involving oviposition is probably notlikely to occur since tansy odour strongly reduces the oviposition in that insect (GABEL& THIERY, 1994).

The aim of this work was to determine the behavioural relevance of differentextraction/fractionation procedures from tansy flowers. This approach is integratedwithin the framework of developing natural odourant baits to trap females, and oneobjective was to propose low cost but efficient baits. From these results, global extractsobtained from the 2 different methods of extractions we have used can be considered asequivalent. Fractionation according to our procedure led to strong decrease of activity.Both EO and E30 elicit a good level of behavioural activity. It will serve in furtherexperiments as standard blends in order to progressively reduce the number of constituents in different mixtures of synthetic molecules by comparing trapping capacitie s onEGVM females in vineyards.

AcknowledgementsWe thank C. Masson (INRA-CNRS, Bures sur Yvette, France) and J. Stockel (INRA, CR

Bordeaux, France) for providing experimental facilities, D. Light (USDA, RC Albany, California, USA) andthe anonymous reviewer for manuscript improvement. We also "give many bisous" to B. Quenet (INRACNRS, Bures sur Yvette, France) for help with statistics.

ReferencesDEMBITSKI, A. D.; KROTOVA, G.!.; YURINA, R.A. & SULEEVA, R., 1984: Composition of the essential oil of

Tanacetum vulgare. Khim. Prir. Soedin., 6: 716-720.GABEL, B., 1992: Tansy flowers attract European grapevine moth females, Lobesia botrana Den. & Schiff.(Lep., Tortricidae). J. Appl. Ent. 113: 153-158.GABEL, B.; THIERY, D.; SUCHY, V.; MARION-POLL, F.; HRADSKY, P. & FARKAS, P., 1992: Floral volatiles of

Tanacetum vulgare L. attractive to Lobesia botrana Den. et Schiff. females. J. Chern. Eco/.,5: 693-701.

GABEL, B. & THIERY, D., 1994: Non-host plant odor (Tanacetum vulgare, Asteracea) affects the reproductive behavior of Lobesia botrana Den. et Schiff. J. Insect Behav., 7 (2) in press.

HOKKANEN, H.; GRANLUND, H.; HUSBERG, G.-B. & MARKKULA, M., 1986: Trap crops used successfully to controlMeligethes aeneus (Col., Nitidulidae), the rape blossom beetle. Ann. Entomol. Fennici, 52: 115-120.

6


7/7

KAISSLlNG, K. E., 1987: R. H Wright lectures on insect olfaction. K. Colbow ed., Simon Fraser Univ. Canadapubl., 75pp.LATHEEF, M. A. & ORTIZ, J. H., 1984: Influence of companion herbs on Phyllotreta cruciferae (Coleoptera:Chrysomelidae) on coliard plants. J. Econ. Entomol., 77: 70-82.LUSTNER, G., 1914: Das Verhalten der Raupen des ein- und zweibindigen Traubenwicklers zu denWeinbergskrautern und anderen Pflanzen. Z. Weinbau und Weinbehandl., 1: 3-35.MARION-POLL, F. & TOBIN T. R., 1991: Software system for detecting spikes superimposed on a fluctuatingbaseline. J. Neurosci. Meth.. 37: 1-6.MAYER, M. S.; MANKIN, R. W. & GRANT, A. J., 1987: Quantitative comparison of behavioral and neurophysiological responses of insects to odorants: inferences about central nervous system processes.

J. Chem. Ecol., 3: 509-531.PERRIN, R. M., 1977: Pest management in multiple cropping systems. Agro-ecosystems, 3: 93-118.SACHS, L., 1984: Applied statistics. A handbook of techniques. New York-Berlin-Heidelberg-Tokyo:Springer.STANTON, M., 1982: Searching in a patchy environment: foodplant selection by Colias eriphyle

P. butterflies. Ecology, 63: 839-853.STOEVA, R., 1982: Les hotes de l'Eudemis (Lobesia botrana SchifL) en Bulgarie. Gradin. Lozar. Nauka,19: 83-89.TAYLOR, T. A., 1977: Mixed cropping as an input in the management of crop pests in tropical Africa. AfricanEnviron., 3: 111-126.THE PHARMACOPOEA BOHEMOSLOVACA, 1987. Editio quarta. Prague: Avicenum Publ., 101-102.THIERY, D. & VISSER, J. H., 1986: Masking of host-plant odour in the olfactory orientation of the coloradopotato beetle. Entomol. expo appl., 41: 165-172.UVAH, I. I. & COAKER, T. H., 1984: Effect of mixed cropping on some insect pests of carrots and onions.Entomol. expo appl., 36: 159-167.VISSER, H., 1979: Electroantennogram responses of the Colorado potato beetle, Leptinotarsa decemplineata, to plant volatiles. Ent. expo & appl., 25: 86-97.WEIR, J. B. DE V., 1960: Significance of the difference between two means when the population varia ncesmay be unequal. Nature, 187: 438.

Manuscript received: 3. 3. 1994

7"

Documents

1994_Entomol_Probl_Gabel_et_al_25_1